From 0aeb6025bdc08444c0c297824e3d77e39ab69203 Mon Sep 17 00:00:00 2001 From: "Bill(Shuzhou) Liu" Date: Thu, 13 Jul 2023 08:28:53 -0500 Subject: [PATCH 1/3] rocm-smi --showevents shows wrong gpuID Use the gpuid returned from the event data instead. Change-Id: I7f286cc105f7ea12985223e603504f0ef3d9724e --- python_smi_tools/rocm_smi.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/python_smi_tools/rocm_smi.py b/python_smi_tools/rocm_smi.py index 5c12624142..2fe68ab518 100755 --- a/python_smi_tools/rocm_smi.py +++ b/python_smi_tools/rocm_smi.py @@ -537,7 +537,7 @@ def printEventList(device, delay, eventList): data = rsmi_evt_notification_data_t(1) rocmsmi.rsmi_event_notification_get(delay, byref(num_elements), byref(data)) if len(data.message) > 0: - print2DArray([['\rGPU[%d]:\t' % (device), ctime().split()[3], notification_type_names[data.event.value - 1], + print2DArray([['\rGPU[%d]:\t' % (data.dv_ind), ctime().split()[3], notification_type_names[data.event.value - 1], data.message.decode('utf8') + '\r']]) def printLog(device, metricName, value=None, extraSpace=False): From b0fe2fbd0762db27c263fdbae46b58dfbfbbb3e4 Mon Sep 17 00:00:00 2001 From: "Galantsev, Dmitrii" Date: Fri, 14 Jul 2023 02:22:45 -0500 Subject: [PATCH 2/3] Add .cache to gitignore Change-Id: Ida03bf1f50704bea44827d7578cd74c1896d4368 Signed-off-by: Galantsev, Dmitrii --- .gitignore | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.gitignore b/.gitignore index 4abf0e47d9..55abee102c 100644 --- a/.gitignore +++ b/.gitignore @@ -12,3 +12,6 @@ python_smi_tools/rsmiBindings.py # Build directory build/ + +# CMake cache +.cache/ From 8fe848d10e485af839fb43501027d2442994021f Mon Sep 17 00:00:00 2001 From: "Galantsev, Dmitrii" Date: Fri, 14 Jul 2023 02:22:50 -0500 Subject: [PATCH 3/3] Fix sys and id tests The following read tests were failing: *.TestIdInfoRead *.TestSysInfoRead 1. *.TestIdInfoRead failed because rsmi_dev_brand_get did not specify dependency on vbios_version. 2. *.TestSysInfoRead failed because the test didn't expect vbios_version to be missing. Which is a new behavior in Aqua Vanjaram. Change-Id: I9ee88a12fcf6cff2032049e2ecdfb2957efb03ab Signed-off-by: Galantsev, Dmitrii --- src/rocm_smi_device.cc | 3 ++- tests/rocm_smi_test/functional/id_info_read.cc | 2 +- tests/rocm_smi_test/functional/sys_info_read.cc | 2 +- 3 files changed, 4 insertions(+), 3 deletions(-) diff --git a/src/rocm_smi_device.cc b/src/rocm_smi_device.cc index 18c96b7f13..554b8c0eb8 100755 --- a/src/rocm_smi_device.cc +++ b/src/rocm_smi_device.cc @@ -379,7 +379,8 @@ static const std::map kDevFuncDependsMap = { {"rsmi_dev_name_get", {{kDevVendorIDFName, kDevDevIDFName}, {}}}, {"rsmi_dev_sku_get", {{kDevDevProdNumFName}, {}}}, - {"rsmi_dev_brand_get", {{kDevVendorIDFName}, {}}}, + {"rsmi_dev_brand_get", {{kDevVendorIDFName, + kDevVBiosVerFName}, {}}}, {"rsmi_dev_vendor_name_get", {{kDevVendorIDFName}, {}}}, {"rsmi_dev_serial_number_get", {{kDevSerialNumberFName}, {}}}, {"rsmi_dev_subsystem_id_get", {{kDevSubSysDevIDFName}, {}}}, diff --git a/tests/rocm_smi_test/functional/id_info_read.cc b/tests/rocm_smi_test/functional/id_info_read.cc index 5c7eea1256..11828feb85 100755 --- a/tests/rocm_smi_test/functional/id_info_read.cc +++ b/tests/rocm_smi_test/functional/id_info_read.cc @@ -162,7 +162,7 @@ void TestIdInfoRead::Run(void) { IF_VERB(STANDARD) { std::cout << "\t**Device Vram Vendor name: " << buffer << std::endl; } - err = rsmi_dev_brand_get(i, nullptr, kBufferLen); + err = rsmi_dev_vram_vendor_get(i, nullptr, kBufferLen); ASSERT_EQ(err, RSMI_STATUS_INVALID_ARGS); } err = rsmi_dev_vendor_id_get(i, &id); diff --git a/tests/rocm_smi_test/functional/sys_info_read.cc b/tests/rocm_smi_test/functional/sys_info_read.cc index 14d1977676..dcff82aa84 100755 --- a/tests/rocm_smi_test/functional/sys_info_read.cc +++ b/tests/rocm_smi_test/functional/sys_info_read.cc @@ -105,7 +105,7 @@ void TestSysInfoRead::Run(void) { err = rsmi_dev_vbios_version_get(i, buffer, 80); if (err != RSMI_STATUS_SUCCESS) { - if (err == RSMI_STATUS_FILE_ERROR) { + if ((err == RSMI_STATUS_FILE_ERROR) || (err == RSMI_STATUS_NOT_SUPPORTED)) { IF_VERB(STANDARD) { std::cout << "\t**VBIOS read: Not supported on this machine" << std::endl;